Literacy and growth: Policy implications of new evidence from PIAAC

Guido Schwerdt         Simon Wiederhold    T. Scott Murray

Educators often dismiss the results of studies, such as PIAAC and PISA, as irrelevant, with no meaningful impact on their priorities or policies.  Observed PIAAC skill differences among countries appear to be relatively small, so are easy to dismiss.

This report summarizes an analysis of PIAAC data by Wiederhold and Schwerdt of the impact that differences in average adult literacy and numeracy skill, and that differences in the distribution of literacy skill by proficiency levels, have had on long-term growth rates of GDP per capita and labor productivity.

It suggests that economic and educational policy makers ignore PIAAC results at their peril. Specifically, the report provides incontrovertible proof that seemingly small differences – as little as 5 points on the 500-point PIAAC proficiency scale in average scores - precipitate quite remarkable increases key indictors of economic performance.

As of 2011, a 1% increase in average literacy scores were was associated with a 3% increase in the growth rate of GDP per capita and a 5% increase in the rate of labor productivity growth. This finding is crucially important for both educational and economic policy, as it suggests that one could, with relatively minor educational investments, double the current rates of economic growth in Canada and the US.

Importantly for policy makers, the impact of literacy skill differences on growth rates have doubled since this analysis was first undertaken using data from the 2003 Adult Literacy and Life Skills Survey (ALL). Literacy is becoming more, rather than less important, as technology and globalization reduce the demand for workers who only read well enough to apply routine procedural knowledge, conventional Level 2 tasks.

Any worker without at least Level 3 literacy will have difficulty generating sufficient value for an employer to hire them.

Unfortunately for policy makers in Canada and the US, the available research seems to suggest that the skill of the adult population has been falling over the past three decades in response to falling scores of PISA aged students and skill loss occurring over the life course.[1]  Clearly, education and economic policy makers should focus their attention on raising average literacy scores of the adult workforce by improving the scores of students leaving the secondary and post-secondary education systems and by investing in literacy skill upgrading for both employed and unemployed adults.

The analysis also reveals a second important finding – that higher proportions of adults with Level 1 and 2 literacy skills reduce growth rates significantly. Thus, investments in literacy skill upgrading that focus on increasing the average skill level by increasing the skill levels of the least skilled would yield both significant economic growth and material reductions in skill-based inequalities in employment, wage rates and incomes. It would make everyone at all skill levels richer, faster. Rarely does one see such a direct confirmation of the truism “rising tides raise all boats”.

Interestingly, the analysis reveals that the proportion of highly literate adults – conventionally Level 4 and 5 - has no impact on relative long-term economic growth rates. This finding should not be interpreted to mean that advanced literacy skills don’t matter to growth but rather that no country is realizing any comparative advantage from the elite end of the skill distribution.

The researchers find that increases in literacy skill lead, rather than lag, increases In economic growth.  More over, the analysis also suggests that economies reach a new steady state after an increase in literacy skill averages rather quickly by economic standards. Specifically, the analysis suggests that it would take between six and nine years to close half of the gap to the new economic steady state i.e. the point at which all of the benefits of higher skill levels have all been realized. Were a way found to increase average literacy skill levels, the OECD economies would reap the benefits relatively rapidly.

These findings support a view that literacy is a fundamental determinant of economic progress, a result that mirrors research undertaken in Canada that documents the positive impact that literacy skill upgrading can have on worker productivity and firm performance[2].  Conducted as a large-scale, randomized controlled trial in the food and accommodation industry, both workers and firms realized a 24% first year rate of return on a 24-point average improvement in their literacy skill level that moved most workers from Levels 1 and 2 to Level 3.

Viewed from a policy perspective, these results reinforce the need for education policy makers to renew their focus on the fundamentals from the beginning of the educational process. More pointedly, educators would be well served to avoid the siren calls of 21st Century and soft skills proponents until they ensure that all children are reading well enough to acquire and apply these skills efficiently.

Finally, the analysis suggests that differences in the average level of women’s skills have had a larger impact on rates of economic growth than men’s. Policy makers could, by raising women’s average literacy levels more rapidly, precipitate even higher levels of growth.

To summarize: Differences in economic growth rates are important because they determine the rate at which societies and economies get wealthier.  Small differences seem to matter a lot. Economic and educational policy makers should take note.

The economy realizes the benefits of higher literacy scores rapidly

[1] See Skills Matter: FURTHER RESULTS FROM THE SURVEY OF ADULT SKILLS OECD 2016, Figure 2.20 • Changes in literacy scores in IALS, ALL and PIAAC surveys Mean literacy proficiency in the International Adult Literacy Survey (IALS),
the Adult Literacy and Lifeskills Survey (ALL) and the Survey of Adult Skills (PIAAC)


[2] See Upskill: A credible test of literacy and Essential Skill Upgrading, SRDC, 2014